MACHINE LEARNING–BASED TOOL WEAR PREDICTION IN ADVANCED MACHINING PROCESSES

Abstract

Tool wear significantly affects surface quality, dimensional accuracy, and productivity in advanced machining processes. Traditional tool wear monitoring methods rely on manual inspection or threshold-based signals, which are often inaccurate and inefficient. This paper proposes a machine learning–based framework for predicting tool wear in advanced machining operations. Sensor data such as cutting force, vibration, acoustic emission, and temperature are utilized to capture tool condition information. Machine learning models are trained to learn complex nonlinear relationships between process signals and tool wear states. Experimental validation demonstrates high prediction accuracy and early detection capability. Comparative analysis shows that the proposed approach outperforms conventional methods. The results confirm that machine learning enables reliable and intelligent tool wear prediction for smart manufacturing environments.